JPH0961386A - Method for detecting photothermally displaced image and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for detecting photothermally displaced image and a photothermally displaced image detecting device in which the internal information for the surface of a sample and its vicinity can be two-dimensionally detected at a high speed with a simple structure. SOLUTION: A plurality of measuring points on a sample are simultaneously excited by a linear intensity modulated light 11, and the thermal expansion displacement generated in each point is simultaneously detected by utilizing the light interference between a linear probe light 18 and a reference light 19, thereby, the photothermal displacement signals in a plurality of measuring points of the sample are simultaneously detected. Since the photothermal displacement signals in the measuring points of the sample can be thus simultaneously detected in parallel, the two-dimensional information for the sample surface and the inner part can be detected at a high speed. A higher speed and a higher sensitivity can be simultaneously realized by the use of a common optical path type interferometer.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a photothermal displacement image detecting method and apparatus for detecting surface and internal information of a sample by utilizing a photoacoustic method or a photothermal displacement method.

[0002]

2. Description of the Related Art In the conventional photothermal displacement method, as shown in FIG. 10, a sample 2 is irradiated with intensity-modulated light 1 and a periodical thermal expansion displacement 4 generated on a sample surface along with a heat wave 3 is detected. , Get inside information of sample 2. The amplitude and phase of the thermal expansion displacement 4 change greatly depending on the thermal conductivity of the surface and the inside. For example, in a region 5 of a semiconductor wafer having crystal defects or damage, the thermal conductivity is lower than that in a normal portion, so that the thermal expansion amplitude is increased and the phase is delayed. Therefore, intensity modulated light 1
The defect region 5 can be identified by measuring the thermal expansion displacement 4 at each point while scanning, using, for example, optical interference. Regarding the above-mentioned prior art, for example, the document “Non-destructive inspection; Vol. 36, No. 10, p. 730 to p. 736 (October 1987)” and “I-E-E-1986”.
Ultra Sonics Symposium; p. 515-5
26 (1986) (IEEE1986 ULTRASO
NICS SYMPO-SIUM; p. 515-526
(1986) ”.

[0003]

Generally, the thermal expansion displacement, which is the object of measurement, is very weak, on the order of sub-nanometers or less, and is buried in noise components such as air fluctuations around the device and stage vibrations in the interference signal. Often In order to extract the amplitude or phase of such a weak periodic component from the interference signal, synchronous detection by a lock-in amplifier is usually used. When the modulation frequency is on the order of several tens of kHz, it is necessary to set the time constant of the lock-in amplifier to about several hundred ms in order to measure the weak signal of the above order. Considering this time constant, it takes an enormous amount of time to measure a two-dimensional region in the conventional photothermal displacement method in which the sample movement and the measurement are repeated point by point. For example, if the time constant is 400 ms,
The measurement time of the area of 256 × 256 points is about 7.3 h.
Becomes

For this reason, the photothermal displacement method has not yet been put to practical use in the field of image measurement, although it is effective for physical property evaluation and measurement of surface layer defects. Achieving a significant increase in speed without reducing sensitivity is the greatest challenge of the photothermal displacement method.

It is an object of the present invention to provide a photothermal displacement image detection method and apparatus capable of two-dimensionally detecting internal information on the surface of a sample and its vicinity with a simple structure.

[0006]

In order to achieve the above object, the present invention irradiates a plurality of measurement points on the surface of a sample with light whose intensity is modulated at a frequency f _E set to be changeable, and the plurality of measurement points are irradiated. The cyclic thermal expansion displacement is generated on the surface of the measurement point of
Other light is radiated to the plurality of measurement points to cause the reflected light to interfere with the reference light whose optical frequency differs by f _B , and the generated interference light is detected by a plurality of photoelectric conversion elements corresponding to the respective measurement points. Detected by the detector, from the detected interference light intensity signal, the interference light intensity change based on the thermal expansion displacement is detected as a photothermal displacement signal at each of the plurality of measurement points, and a plurality of samples of the sample are obtained from the photothermal displacement signal. It detects the surface and internal information of the measurement point.

In order to achieve the above object, the present invention provides a method of irradiating a plurality of measurement points on the surface of a sample with intensity-modulated light, in which the intensity-modulated light forms a continuous linear shape on the sample. It is configured as a beam.

In order to achieve the above object, the present invention has a structure in which interference light is integrated and detected by the detector.

Further, in order to achieve the above object, the present invention is configured to use a detector in which an interference light intensity signal is output as a one-dimensional signal in time series from a plurality of photoelectric conversion elements. .

Further, in order to achieve the above object, according to the present invention, the frequencies f _S , f _B and f _E are 8p: 8pu + 1: 8p.
v-1, or 8p: 8pu-1: 8pv + 1 (p,
u, v: an arbitrary integer other than 0) and 1 / f _S for each photoelectric conversion element of the detector under the control of a constant integer ratio.
From a plurality of integrated detection data which are integrated and detected a plurality of times in the time period of, the interference light intensity change based on the thermal expansion displacement is detected as a photothermal displacement signal for each of the plurality of measurement points.

In order to achieve the above-mentioned object, the present invention irradiates the reference light at a position approximately distant by a thermal diffusion length or more from the irradiation point of the intensity-modulated light on the sample. It is what

In order to achieve the above-mentioned object, the present invention provides that the intensity modulation frequency f _E is set so that the thermal diffusion length based on the photothermal effect or the photoacoustic effect is almost the same as the depth of the internal interface to be measured of the sample. Alternatively, the length is set to exceed that.

In order to achieve the above object, the present invention comprises, as constituent elements of the apparatus, a light source, intensity modulating means for intensity-modulating light from the light source at a frequency f _E set to be changeable, and Excitation means for irradiating a plurality of measurement points on the sample surface with intensity-modulated light to generate cyclic thermal expansion displacement on the surface of the plurality of measurement points, and irradiating the plurality of measurement points with other light. Interference light generation means for causing the reflected light to interfere with the reference light whose optical frequency differs by f _B, and the generated interference light has a conjugate relationship with the sample surface and a plurality of photoelectric conversion points corresponding to the respective measurement points. Interference light detection means for detecting with a detector composed of a conversion element, and from the detected interference light intensity signal, the interference light intensity change based on the thermal expansion displacement is detected as a photothermal displacement signal at each of the plurality of measurement points. , Multiple measurements of the sample from the photothermal displacement signal It is provided with information detecting means for detecting the surface and internal information of the point, and storage means and display means for the surface and internal information.

[0014]

In the photothermal displacement image detection device, the light of which the intensity is modulated at the frequency f _E set to be changeable is applied to a plurality of measurement points on the sample surface, so that the surfaces of the plurality of measurement points undergo cyclic thermal expansion. Along with being able to generate displacement, other light is applied to a plurality of measurement points to cause the reflected light to interfere with reference light whose optical frequency differs by f _B , and the generated interference light corresponds to a plurality of measurement points. Detected by a detector consisting of photoelectric conversion elements, from among the detected interference light intensity signal, by detecting the interference light intensity change based on the thermal expansion displacement as a photothermal displacement signal for each of the plurality of measurement points The surface and internal information of a plurality of measurement points of the sample can be extracted almost at the same time, and the photothermal displacement image can be detected much faster than the conventional method.

Further, as a method of irradiating a plurality of measurement points on the surface of the sample with the intensity-modulated light, the intensity-modulated light is configured as a beam having a continuous linear shape on the sample, so that it is remarkably different from the conventional method. This makes it possible to detect a photothermal displacement image at high speed.

Further, the photothermal displacement image can be detected at high speed and with high accuracy by adopting a configuration in which the interference light is integrated and detected by the detector.

Further, by using a detector in which the interference light intensity signal is outputted as a one-dimensional signal in time series from a plurality of photoelectric conversion elements, it is possible to detect a photothermal displacement image at a much higher speed than the conventional method. It will be.

Further, in order to achieve the above object, according to the present invention, the frequencies f _S , f _B and f _E are 8p: 8pu + 1: 8p.
v-1, or 8p: 8pu-1: 8pv + 1 (p,
u, v: an arbitrary integer other than 0) and 1 / f _S for each photoelectric conversion element of the detector under the control of a constant integer ratio.
From a plurality of integral detection data integrated and detected a plurality of times in the time period of, by detecting the interference light intensity change based on the thermal expansion displacement as a photothermal displacement signal for each of the plurality of measurement points, the conventional method This makes it possible to detect photothermal displacement images with significantly higher speed, higher sensitivity, and higher accuracy than in.

Further, by irradiating the reference light at a position approximately distant from the irradiation point of the intensity-modulated light by a thermal diffusion length or more, the speed is much higher than that of the conventional method, and the sensitivity and the accuracy are high. The photothermal displacement image can be detected.

Further, the intensity modulation frequency f _{E is set so} that the thermal diffusion length based on the photothermal effect or the photoacoustic effect is approximately the same as or longer than the depth of the measured internal interface of the sample. This makes it possible to inspect an arbitrary internal interface and an arbitrary depth of the sample.

[0021]

Embodiments of the present invention will be described with reference to FIGS.

The factor of decreasing the detection speed in the conventional method is that point scanning detection and lock-in amplifier, that is, synchronous detection are used. An object of the present invention is to realize a significant speedup while maintaining the same sensitivity as the conventional method. Before concrete description of the embodiments, a basic concept and a basic principle for speeding up will be described.

The basic concept is shown in FIG. First, in order to significantly speed up the conventional point scanning detection method, we propose a parallel pumping / parallel optical interference method using a linear beam that simultaneously pumps / detects multiple points on a sample. By implementing this parallel optical interferometry with a common optical path type interferometer, noise components in the interference signal due to air fluctuations and vibrations during stage scanning can be significantly reduced. As a result, it becomes possible to continuously scan the stage, which was conventionally stopped one by one during measurement, and the speed can be further improved.

In the parallel optical interferometry, it is necessary to use a one-dimensional sensor composed of a plurality of light receiving elements corresponding to the measurement point for detecting the interference light, and the detection signal cannot be directly input to the lock-in amplifier. Can not. Therefore, we propose a new thermal expansion component extraction processing method and sampling frequency conversion method instead of the lock-in amplifier, that is, the synchronous detection. In this processing method, the effect of reducing the noise component due to the adoption of the common optical path type interferometer is utilized, and the number of data used for signal processing is reduced, thereby significantly reducing the detection time per point.

First, the principle of parallel excitation / parallel optical interferometry will be described. As shown in FIG. 1, when the intensity-modulated light 11 having the frequency f _E is reflected by the dichroic mirror 12 and is linearly condensed on the surface of the sample 10, and a plurality of consecutive points are excited in parallel, linearly periodical light is generated. Thermal expansion displacement 14 occurs. The amplitude and phase of the thermal expansion displacement 14 at each point change according to the distribution of thermal conductivity on the sample surface and inside. Excitation unit 1
3, the same linear probe laser light 18 separated from the laser light 15 by the birefringent element 17, and the reference light 19 whose optical frequency is slightly different from that of the probe light 18 in the vicinity thereof.
Are radiated through the beam splitter 16 and the dichroic mirror 12, respectively. A linear thermal expansion displacement distribution 14 is superimposed on the probe light reflected from the excitation unit 13 as a one-dimensional optical phase distribution. Birefringent element 17 for both reflected light
And the heterodyne interference is combined, and this optical phase distribution is collectively detected by the one-dimensional CCD sensor 21 as the one-dimensional spatial distribution of the heterodyne interference light 20.

By adopting a common optical path type interferometer in which the probe light 18 and the reference light 19 pass through almost the same optical path, the optical phase difference due to the fluctuation of air generated in both optical paths and the vibration during the stage scanning is canceled. It is possible to significantly reduce the fluctuation of the interference signal. This enables continuous scanning of the stage.

Next, the sampling frequency conversion method will be described. In FIG. 1, the amplitude of the thermal expansion displacement 14 generated at one point on the surface of the sample 10 is A, the phase is θ, and the probe light 1
8 is λ, its reflection intensity is I _S , the reflection intensity of the reference light 19 is I _r , the optical frequency difference between the probe light 18 and the reference light 19, that is, the heterodyne beat frequency is f _B , and
Let φ be the phase difference between both optical paths caused by the unevenness of the surface of the sample 10. In general, the thermal expansion amplitude A is on the order of sub-nanometers or less. Therefore, if λ = 632.8 nm, for example, A << λ holds and the intensity of the heterodyne interference light 20 incident on one pixel with the one-dimensional CCD sensor 21. I
(T) can be approximated by the following equation.

[0028]

[Equation 1]

The first term is the DC component, the second term is the carrier component of the heterodyne beat frequency f _B , and the third term is the thermal expansion amplitude A.
And a sideband component of frequency f _B + f _E having information of phase θ,
The fourth term is the frequency f which also has the information of the amplitude A and the phase θ.
It is a sideband component of _B- f _E. The third or fourth term is the information component to be obtained.

As shown in FIG. 3, the interference light 20 is integrated and converted at a constant cycle due to the accumulation effect of the CCD sensor 21, and an interference signal 25 is output for each pixel column by line scanning. Therefore, it is impossible to directly input the output signal to the lock-in amplifier and extract the thermal expansion amplitude A and the phase θ by synchronous detection.

Therefore, conversely, by utilizing the accumulation effect of the CCD sensor 21, the relation between the accumulation frequency f _S , the beat frequency f _B and the intensity modulation frequency f _E is 8p: 8pu + 1: 8p.
v-1, or 8p: 8pu-1: 8pv + 1 (p,
u and v are arbitrary integers other than 0). For example, p = 1,
If v = u = 2, then f _S : f _B : f _E = 8: 17: 15
Therefore, the heterodyne interference signal S (i) from a specific one pixel in the i-th line scan is expressed by Formula 2 by the accumulation (integration) effect of the sensor and the sampling effect for each line scan.

[0032]

[Equation 2]

[0033]

(Equation 3)

That is, as shown in FIG. 3, f _B + f _E is f
_{Since it} is an integral multiple of _S , this thermal expansion component 23 is 1 / f _S
After being integrated and under-sampled in a period of, the signal is converted into a DC component (black plot in the figure) and removed from the output signal. On the other hand, the beat component 22 of the frequency f _B is
Similarly, the line scan i, that is, the component of frequency 1/8 with respect to the period of 1 / f _S , and the thermal expansion component 24 of f _B -f _E are converted into components of frequency 1/4, respectively. It

Next, the principle of extracting the thermal expansion component by the Fourier coefficient will be described. The discrete Fourier series expansion for a periodic signal and its Fourier coefficient are given by the following equation.

[0036]

(Equation 4)

[0037]

(Equation 5)

[0038]

(Equation 6)

[0039]

(Equation 7)

[0040]

(Equation 8)

[0041]

[Equation 9]

[0042]

(Equation 10)

By substituting each equation into the equations 11 and 12, the amplitude A and the phase θ of the thermal expansion displacement can be obtained.

[0044]

[Equation 11]

[0045]

(Equation 12)

In the denominator of equation 11, the reflection intensities I _S and I _r of the probe light 18 and the reference light 19 (from the common optical path, I _S =
I _r ) is included, and the influence of the reflectance distribution on the surface of the sample 10 is corrected. The second term of Expression 12 is a term for correcting the phase change due to the unevenness of the surface of the sample 10 obtained from the beat component. As described above, by converting the beat component and the thermal expansion component so that the frequency ratio becomes 1: 2 by the sampling frequency conversion, the thermal expansion component is accurately extracted from the interference signal only by simple addition and subtraction processing. The point that can be achieved is a major feature of this processing method. In the signal processing circuit to be described later, by utilizing this feature, by sequentially writing / reading the output signal 25 from the CCD sensor 21 to / from the one-dimensional memory and adding / subtracting processing,
The derivation of each Fourier coefficient shown in Expressions 7 to 10 is executed in real time.

A specific embodiment of the present invention will be described with reference to FIGS.

FIG. 4 shows the overall construction of the photothermal displacement image detection device. This apparatus is composed of an optical system portion including a parallel excitation optical system 60 and a parallel interference optical system 61, a signal input circuit 55, a signal processing circuit 56, a signal control circuit 57, a computer 58, a display 59, and a stage system 54. .

As the excitation light source, a high power semiconductor laser 30 having an output of 2 W and a wavelength of 799 nm having a linear light emitting region of 1.5 × 200 μm was used. The intensity of the light emission pattern was directly modulated by modulating the drive current with a rectangular wave of frequency f _E.

When this light emission pattern is imaged on the sample 37 by the collimating lens 31, the relay lenses 32 and 33, and the objective lens (magnification 20 ×, NA 0.4) 36, the intensity-modulated linear beam is naturally formed. 34 is formed, and a plurality of continuous points are excited in parallel to generate a linear cyclic thermal expansion displacement. The linear beam 34 was approximately 2 × 210 μm on the sample 37.

As the probe light source, a frequency-stabilized He-Ne laser 39 having a wavelength of 632.8 nm was used. The linearly polarized light beam emitted from the laser is incident on the optical frequency shifter 40, and is separated into two polarized light components 41 which have slightly different optical frequencies and are orthogonal to each other. This dual frequency orthogonal polarization component 41
Is expanded by the beam expander 42, reflected by the mirror 43, transmitted through the cylindrical lens 44 and the beam splitter 45, and then, as shown in FIG. 5, calcite 46a, 46b.
Are separated in parallel by the birefringence characteristic of the Savart plate 46, and after passing through the relay lens 47 and the dichroic mirror 35, the objective lens 36 causes one of the probe light 48 to be the excitation unit 38 and the other to be the reference light 49 to be adjacent to the excitation light.
Each is irradiated as a linear beam.

Both reflected lights are recombined by the Savart plate 46, reflected by the dichroic mirror 45, transmitted through the imaging lens 50, and then heterodyne polarization interference is caused by the polarizing plate 51, and the linear optical phase distribution due to the thermal expansion displacement interferes. Light 52
Are collectively detected by the one-dimensional CCD sensor 53. The number of pixels of the sensor is 256, and the pixel size is □
The detection visual field on the sample 37 was set to about 166 μm by using the objective lens 36 having a magnification of 20 ×.

The interval between the probe light 48 and the reference light 49 is set to 20 μm, which exceeds the obtained thermal diffusion length of 18 μm, when the measurement sample is a silicon wafer and the modulation frequency f _E is 88.235 kHz.

FIG. 6 shows the overall configuration of the signal input circuit 55, the signal processing circuit 56, and the signal control circuit 57.

In the signal input circuit 55, the interference signal output from the one-dimensional CCD sensor 53 is adjusted by offset / gain adjustment (not shown), and then the AD converter 70 outputs 14 bits.
AD conversion to digital data of.

[0056] The signal processing circuit 56, ALU (A rith
consists metic L ogic U nit) 71,1-dimensional memory 72, dimensional memory 73 and the computer 58, C
For the interference signal from the CD sensor 53, several 7 to several 10
The addition and subtraction processing of is repeatedly executed, and the processing result is written in the two-dimensional memory 73.

That is, in the ALU 71, with respect to the read data x from the one-dimensional memory 72 corresponding to the 256 pixels of the sensor 53 and the interference signal y from the signal input circuit 55,
Addition processing z = x + y or subtraction processing z = for each pixel
xy is executed and the calculation result z is written to the same address in the one-dimensional memory 72. After the above processing is repeatedly executed, the contents of the one-dimensional memory 72 are written in the two-dimensional memory 73. This line data corresponds to the addition / subtraction processing data for one line of the measurement area. By performing the above processing while continuously scanning the stage 54, the addition / subtraction processing in the two-dimensional area is completed.

The signal processing circuits 56 are arranged in five sets and are configured to execute a total of five types of arithmetic processing in parallel, that is, the addition / subtraction processing of the equations 7 to 10 and the addition processing for obtaining the DC component. After the processing is completed, a two-dimensional image of the thermal expansion amplitude A and the phase θ is calculated by the computer 58 from the Fourier coefficient data and the DC component data stored in each two-dimensional memory 73 based on the equations 11 and 12, and the display 59 is displayed. To display.

The ability to separate and extract the weak thermal expansion component is the frequency condition for executing the sampling frequency conversion, that is, 8p: 8pu + 1: 8pv-1, or 8p: 8p.
u-1: 8pv + 1 (p, u, v are arbitrary integers other than 0)
Depends on the setting accuracy of. Therefore, the signal control circuit 57
So as to centralize the generation of the control signals adopts a PLL (P hase L ocked L oop ) circuit that generates and to stabilize the frequency and phase. The signal control circuit 57 (PLL circuit) generates the accumulation time control signal (f _S ) and the excitation intensity modulation signal (f _E ) using the heterodyne beat signal of the frequency f _B generated by the parallel interference optical system 61 as a reference signal. , Respectively to the one-dimensional CCD sensor 53 and the semiconductor laser drive driver 77. Modulation frequency f _E
Is variable in the range of 1.786 to 88.235 kHz.

The experimental results in this example are shown below.
Since the beat frequency f _{B is set} to 100 kHz and the sampling frequency conversion is executed, the modulation frequency f _E of the excitation light is 88.235 kHz, the storage frequency f _S of the one-dimensional CCD sensor 53 is 47.059 kHz, and the clock frequency is The frequency f _C was set to 14.118 MHz.

FIG. 7 shows the relationship between the detection time and the minimum detected thermal expansion amplitude. The standard deviation of the thermal expansion amplitude distribution for a uniform Al mirror sample was determined, and this was taken as the noise amplitude, and S /
The amplitude value with N ≧ 1 was defined as the minimum detected thermal expansion amplitude. It can be seen that the detectable thermal expansion amplitude value is inversely proportional to the 1/2 power of the detection time. The minimum detected thermal expansion amplitude was 7.9 pm at a detection time of 40 μs / pixel (hereinafter, referred to as high-sensitivity mode) and 26 pm at 4 μs / pixel (the same, high-speed mode).

FIG. 8 shows the relationship between the crystal defect density (in impurity concentration conversion) generated in the silicon wafer due to etching damage and the thermal expansion amplitude. It can be seen that as the defect density increases, the thermal conductivity decreases and the thermal expansion amplitude increases. Considering the noise amplitude, from the change of the thermal expansion amplitude value to the change of the defect density to the defect density of the order of about 10 ⁹ defects / cm ^{2 in} the high sensitivity mode, and about 10 in the high speed mode.
It is possible to discriminate up to a defect density of about ¹⁰ / cm ² .

FIG. 9 shows an example of imaging the crystal defect distribution of a silicon wafer. The sample is partially Ar on the wafer.
Ion acceleration energy 300 keV, current density 10 m
A crystal defect was locally formed by implanting 10 ¹⁵ pieces / cm ² under the condition of A / cm ² . In the thermal expansion amplitude image shown in FIG. 9A, it can be seen that the defect region where the surface reflectance hardly changes as shown in FIG. This is because the thermal conductivity is lowered in the defect area, so that the surface temperature is increased and the thermal expansion displacement is increased. The detection time at this time is
4μs / pixel in high-speed mode, 0.26s per screen
And about 1 / 100,0 of the conventional method based on point scanning detection
It was 00. No difference in horizontal and vertical resolution was observed with parallel excitation.

As described above, according to the present embodiment, the parallel excitation / excitation using the linear beam is used instead of the point scanning method in which excitation and detection are repeated point by point as in the conventional photothermal displacement method.
Two-dimensional information on the sample surface and inside can be detected at high speed by exciting multiple measurement points at the same time by parallel optical interferometry and simultaneously detecting the thermal expansion displacement generated at each point by optical interference. Become.

Further, by using the sampling frequency conversion method utilizing the accumulation effect of the CCD sensor instead of the synchronous detection method such as the conventional lock-in amplifier, the thermal expansion signal from the interference signal detected by the CCD sensor can be changed. Extraction becomes possible, and high-speed detection and high-sensitivity detection can be achieved at the same time, together with parallel excitation / parallel optical interferometry.

Further, by configuring the parallel optical interferometry with a common optical path type interferometer, the effects of fluctuations in the optical path and stage vibrations can be canceled out, and continuous scanning of the sample becomes possible, resulting in significant speedup and high sensitivity. Will be possible.

Further, since the thermal expansion amplitude is obtained by correcting the influence of the reflectance distribution on the sample surface, and the thermal expansion phase is obtained by correcting the uneven distribution on the sample surface, high accuracy,
Highly sensitive detection is possible.

Whether the thermal diffusion length based on the photothermal effect or the photoacoustic effect is the same as the depth of the internal interface to be inspected,
Alternatively, by setting the intensity modulation frequency of the excitation light so that the length becomes longer than that, it is possible to inspect an arbitrary internal interface and an arbitrary depth.

As described above, the photothermal displacement method can be applied to the two-dimensional image measurement at a practical level, which was difficult with the conventional method, and a new application field utilizing the high-sensitivity detection capability originally possessed by the photothermal displacement method is provided. spread. That is, two-dimensional evaluation / measurement in various fields such as crystal defects of semiconductor devices, in-plane distribution of ion dose amount, peeling of wiring layer of electronic circuit board, in-plane distribution of crystal defects of surface emitting laser and impurities. It will be possible.

[0070]

According to the present invention, two-dimensional information on the sample surface and inside can be detected at high speed by simultaneously exciting a plurality of measurement points and simultaneously detecting the thermal expansion displacement generated at each point by optical interference. It has a great effect that it becomes possible to do.

Further, by using the sampling frequency conversion method utilizing the accumulation effect of the CCD sensor, the CCD
The thermal expansion signal can be extracted from the interference signal detected by the sensor, and it is possible to achieve high-speed detection and high-sensitivity detection simultaneously with the parallel excitation / parallel optical interference method.

Further, by configuring the parallel optical interferometry with a common optical path type interferometer, the effects of optical path fluctuations and stage vibrations can be canceled out, and continuous scanning of the sample becomes possible, resulting in a significant increase in speed and sensitivity. Has the effect that it becomes possible. .

Further, since the thermal expansion amplitude is obtained by correcting the influence of the reflectance distribution on the sample surface, and the thermal expansion phase is obtained by correcting the uneven distribution on the sample surface, high accuracy,
It has the effect of enabling highly sensitive detection.

Whether the thermal diffusion length based on the photothermal effect or the photoacoustic effect is the same as the depth of the internal interface to be inspected,
Alternatively, by setting the intensity modulation frequency of the excitation light so that the length exceeds it, it is possible to inspect an arbitrary internal interface and an arbitrary depth.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic principle of parallel excitation / parallel optical interferometry in the present invention.

FIG. 2 is a diagram showing a basic concept of speeding up in the present invention.

FIG. 3 is a diagram showing a basic principle of a sampling frequency conversion method according to the present invention.

FIG. 4 is a perspective view showing an overall configuration of a photothermal displacement image detection device according to an embodiment of the present invention.

FIG. 5 is a diagram showing a configuration of a common optical path type interferometer of the photothermal displacement image detection device.

FIG. 6 is a diagram showing an overall configuration of a signal input circuit, a signal processing circuit, and a signal control circuit of the photothermal displacement image detection device.

FIG. 7 is a diagram showing the relationship between the detection time and the minimum detected thermal expansion amplitude measured in the example of the present invention.

FIG. 8 is a diagram showing a relationship between a crystal defect density (converted to an impurity concentration) generated in a silicon wafer and a thermal expansion amplitude measured in an example of the present invention.

FIG. 9 is a diagram showing an example of imaging a crystal defect distribution of a silicon wafer measured in an example of the present invention.

FIG. 10 is a diagram showing a configuration of a conventional photothermal displacement method.

[Explanation of symbols]

1, 11, 34 ... Intensity modulated light, 4, 14 ... Thermal expansion displacement,
15, 41 ... Laser light, 17 ... Birefringent element, 18 ... Probe light, 19 ... Reference light, 20, 52 ... Interference light, 21, 5
3 ... One-dimensional CCD sensor, 30 ... Semiconductor laser, 36 ...
Objective lens, 39 ... He-Ne laser, 40 ... Optical frequency shifter, 44 ... Cylindrical lens, 46 ... Savart plate, 55 ... Signal input circuit, 56 ... Signal processing circuit, 57 ...
Signal control circuit, 58 ... Calculator, 71 ... ALU, 72 ... 1
Dimensional memory, 73 ... Two-dimensional memory.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Irario Kobayashi Haruomi, 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Stock Engineering Co., Ltd.

Claims (14)

[Claims] 1. A plurality of measurement points on a sample surface are irradiated with light whose intensity is modulated at a frequency f _E set to be changeable to generate cyclic thermal expansion displacements on the surfaces of the plurality of measurement points. Other light is radiated to the plurality of measurement points to cause the reflected light to interfere with the reference light whose optical frequency differs by f _B , and the generated interference light is detected by a plurality of photoelectric conversion elements corresponding to the respective measurement points. Detected by the detector, from the detected interference light intensity signal, the interference light intensity change based on the thermal expansion displacement is detected as a photothermal displacement signal at each of the plurality of measurement points, and a plurality of samples of the sample are obtained from the photothermal displacement signal. A photothermal displacement image detection method characterized by detecting the surface and internal information of a measurement point. 2. The method of irradiating a plurality of measurement points on the sample surface with the intensity-modulated light is realized by forming the intensity-modulated light as a beam having a continuous linear shape on the sample. The photothermal displacement image detection method according to claim 1, which is characterized in that: 3. The photothermal displacement image detection method according to claim 1, wherein the interference light is integrated and detected by the detector. 4. The photothermal displacement image detection method according to claim 1, wherein the interference light intensity signal from the detector is output as a one-dimensional signal in time series from a plurality of photoelectric conversion elements. 5. The frequencies f _S , f _B and f _E are 8p: 8pu +.
1: 8pv-1, or 8p: 8pu-1: 8pv + 1
One (1) for each photoelectric conversion element of the detector under the control of a constant integer ratio of (p, u, v: any integer other than 0).
It is possible to detect a change in interference light intensity based on the thermal expansion displacement as a photothermal displacement signal for each of the plurality of measurement points from a plurality of integral detection data that are integrated and detected a plurality of times in a time period of / f _S. The photothermal displacement image detection method according to claim 3, characterized in that 6. The photothermal displacement according to claim 1, wherein the reference light is radiated on the sample at a position approximately distant from the irradiation point of the intensity-modulated light by a thermal diffusion length or more. Image detection method. 7. The intensity modulation frequency f _E is set so that the thermal diffusion length based on the photothermal effect or the photoacoustic effect is substantially the same as or longer than the depth of the measured internal interface of the sample. The photothermal displacement image detection method according to claim 1, wherein 8. A light source, intensity modulation means for intensity-modulating the light from the light source at a frequency f _E set to be changeable, and the intensity-modulated light is applied to a plurality of measurement points on a sample surface,
Excitation means for generating periodic thermal expansion displacements on the surfaces of the plurality of measurement points, and other light irradiating the plurality of measurement points so that the reflected light interferes with the reference light whose optical frequency differs by f _B. Interference light generation means, interference light detection means for detecting the generated interference light with a detector having a conjugate relationship with the sample surface and comprising a plurality of photoelectric conversion elements corresponding to the respective measurement points, and the detection From the interference light intensity signal, the interference light intensity change based on the thermal expansion displacement is detected for each of the plurality of measurement points as a photothermal displacement signal, and the surface and internal information of the plurality of measurement points of the sample from the photothermal displacement signal. A photothermal displacement image detection device, comprising: an information detection unit for detecting the above; and a storage unit and a display unit for the surface and internal information. 9. The means for irradiating the plurality of measurement points on the sample surface with the intensity-modulated light is realized by configuring the intensity-modulated light as a beam having a continuous linear shape on the sample. The photothermal displacement image detection device according to claim 8. 10. A photothermal displacement image detecting apparatus according to claim 8, wherein the detector of said interference light detecting means is constituted by a storage type photoelectric conversion element. 11. The photothermal displacement image detection device according to claim 8, wherein the interference light intensity signal from the detector is output as a one-dimensional signal in a time series from a plurality of photoelectric conversion elements. 12. The frequencies f _S , f _B and f _E are set to 8p: 8pu.
+1: 8pv-1, or 8p: 8pu-1: 8pv +
1 (p, u, v: an arbitrary integer other than 0) is generated in a state of being controlled to a constant integer ratio, and is supplied to a necessary portion with control signal generating means, and each photoelectric conversion element of the detector, From a plurality of integral detection data integrated and detected a plurality of times at a time period of 1 / f _S, the interference light intensity change based on the thermal expansion displacement is detected as a photothermal displacement signal for each of the plurality of measurement points, 11. The photothermal displacement image detection device according to claim 10, further comprising information detecting means for detecting the surface and internal information of a plurality of measurement points of the sample from the photothermal displacement signal. 13. The photothermal displacement image detecting apparatus according to claim 8, wherein the reference light is irradiated at a position approximately distant from the irradiation point of the intensity-modulated light by a thermal diffusion length or more. 14. The intensity modulation frequency f _E is set so that the thermal diffusion length based on the photothermal effect or the photoacoustic effect is substantially the same as or longer than the depth of the measured internal interface of the sample. 9. The photothermal displacement image detection device according to claim 8, wherein: